Texture Analysis for Postmenopausal Osteoporosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Texture Analysis for Postmenopausal Osteoporosis (NCT00145977), a non-randomized na study. The primary outcome is Changes in Lumbar Spine BMD +/- Treatment With Alendronate.

2. Clinical Question

3. Trial Design

The registry describes a parallel, non-randomized study with no masking. It lists an enrollment of 36 and identifies the University of Chicago as sponsor. The recorded study start is 2001-07, primary completion is 2009-12, and status is completed. The listed conditions are osteoporosis and osteopenia.

Participants were assigned to an experimental group or a control group, but the record does not describe random allocation. The experimental group received alendronate 70 mg once weekly as well as calcium citrate and vitamin D. The control group received calcium citrate and vitamin D without alendronate. Because allocation was non-randomized, observed group differences could reflect differences between the people in each group, as well as any difference associated with alendronate. The registry does not report how group assignment was made or whether the groups were comparable at baseline.

The primary outcome was percent change in lumbar spine bone mineral density from baseline to month 24. Secondary outcomes included bone mineral density at the heel, femoral neck, and total hip, along with radiographic texture analysis features assessed from baseline to month 24. The study description identifies computerized analysis of heel x-ray images as an experimental method for examining bone quality.

The registry lists ten outcome measures, but it does not provide the numerical primary or secondary outcome results in the record described. It also does not report a statistical analysis plan, details of recruitment or follow-up, or a rationale for the enrollment target. Those omissions limit assessment of how the study was conducted and how its results should be interpreted.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Three missing values in Alendronate group and five missing values in control group were not imputed.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Changes in Lumbar Spine BMD +/- Treatment With Alendronate, measured over Baseline to Month 24.

The primary endpoint is percent change in lumbar spine bone mineral density from baseline to month 24. Bone mineral density is a quantitative measure of bone at a specified anatomical site. Expressing change as a percentage relates the change to the starting measurement, which can help describe change across participants whose initial measurements differ. The registry specifies that the endpoint is summarized as a mean with standard deviation.

The endpoint is used to assess change over the study period in the lumbar spine measurement, in the context of the study's comparison of groups receiving calcium citrate and vitamin D with or without alendronate. It is distinct from the experimental heel radiographic texture analysis: bone mineral density and image-derived texture features are separate measures, and one cannot be substituted for another without evidence.

The registry record does not report the endpoint's group means, standard deviations, between-group estimate, or confidence interval. Therefore, the magnitude and direction of change, and the precision of any comparison, are not available from the record. It also does not report enough information to determine whether the endpoint was analyzed using an adjusted or unadjusted comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry defines the primary measure as mean percent change in lumbar spine bone mineral density from baseline to month 24, with standard deviation as the dispersion measure. A mean describes the average observed value in a group. A standard deviation describes how spread out individual values are around that average; it is not the same as the uncertainty in the estimated mean. The registry does not report the actual means or standard deviations for the primary endpoint.

A comparison of group means would describe how the recorded endpoint differs between the alendronate group and the control group. The estimate is most informative when accompanied by an interval of uncertainty, such as a confidence interval. In general, a confidence interval gives a range of values compatible with the data under the assumptions of the analysis. A narrower interval indicates greater precision, while a wider interval indicates greater uncertainty. No primary endpoint comparison estimate or confidence interval is reported in the registry record.

A p-value, when reported, describes how compatible the observed data are with a specified null model under the assumptions used for the test. It does not measure the size or clinical importance of an effect, and it does not establish that a treatment caused a difference. The registry record does not report a primary endpoint p-value or identify a test used for that endpoint. A statistical analysis plan, including any adjustment strategy, is not reported in the registry record.

Allocation is an important part of interpreting a group comparison. In a randomized study, chance assignment can help balance both known and unknown participant characteristics across groups, although it does not guarantee exact balance. Here, allocation is listed as non-randomized. In general, this means that group differences may be associated with factors that also influence the outcome. Statistical adjustment can address measured differences under suitable assumptions, but it cannot automatically remove confounding from unmeasured factors. The record does not state whether adjustment was used or which variables might have been considered.

The endpoint is a change from baseline. In general, change scores describe how an individual's measurement differs from their own starting value. The analysis approach matters: comparing group change scores and comparing follow-up values with adjustment for baseline are different approaches, each with assumptions and implications. The registry gives the endpoint definition but does not report the model or method used to compare groups. It is therefore not possible to identify the precise estimand or reproduce the analysis from the record.

Missing data are relevant because the registry notes that three values in the alendronate group and five values in the control group were not imputed. An analysis without imputation uses the available recorded endpoint values rather than filling in missing ones. In general, this can produce a biased estimate if the availability of a measurement is related to the outcome or to factors associated with it. Imputation is not automatically preferable: its validity depends on assumptions about why data are missing and on the method used. The record does not report reasons for missingness, a sensitivity analysis, or further details about the analysis population.

The registry lists a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The illustration is described as using the stated fraction of the observed effect as the effect to detect. Such calculations are prospective planning scenarios: they indicate the sample size that would be needed under stated assumptions for a future study. They are not observed power and do not establish whether the completed study was adequately powered. The record does not report the numerical planning results or the trial's own design justification. The observed effect used for the illustration cannot be evaluated from the registry's absent endpoint results.

General sample-size calculations depend on the target effect, outcome variability, significance level, allocation, and chosen power. If assumptions change, the calculated sample size can change. A planning calculation also cannot correct the interpretive limits created by non-randomized allocation, missing outcome values, or unreported analysis choices. The registry record does not report interim analyses, a prespecified statistical plan, or details about multiplicity control.

Finally, the study lists multiple secondary outcomes, including bone density at other sites and multiple radiographic texture features. When many outcomes are examined, the chance of finding an apparently notable result by chance can increase. In general, multiplicity procedures or a clear hierarchy of outcomes can help manage that issue, but the registry record does not describe such procedures. Without outcome estimates and analysis details, the texture measures cannot be evaluated for statistical reliability or their relationship to bone mineral density.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Alendronate", "Control")
n   <- c(14, 8)
m   <- c(5.28, -1.48)
sp  <- c(10, 4.56)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Alendronate  14  5.280  10.000  -0.494  11.054
CARM  Control  8  -1.480  4.560  -5.292  2.332

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and several secondary outcomes, including bone mineral density at the heel, femoral neck, and total hip, and multiple heel radiographic texture features. The listed outcomes cover more than one type of measure and anatomical site. The record does not report whether a hierarchy among secondary outcomes was prespecified or whether any adjustment for multiple comparisons was planned. Without that information, findings across multiple outcomes would need cautious interpretation: in general, testing many outcomes can increase the chance of results that appear notable by chance.

For the primary endpoint, the registry notes that three values in the alendronate group and five values in the control group were missing and were not imputed. It does not report the reasons for missingness, the timing of missing measurements, or whether the missing values arose for similar reasons in each group. These details matter because missingness can affect which participants contribute to the analysis and can make the observed results less representative of the intended population.

The record does not describe the analysis population, the handling of missing secondary outcomes, or sensitivity analyses exploring alternative assumptions. It also does not report whether the primary analysis adjusted for baseline values or other participant characteristics. These omissions prevent a full assessment of the robustness of the reported endpoint analysis; in this case, the numerical endpoint results themselves are not reported either.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The registry reports no serious adverse events among the participants listed at risk in either group: the alendronate group has 0 serious events among 19 at risk, and the control group has 0 among 17 at risk. These are the serious-event counts recorded for the study groups. They do not establish that no serious events occurred outside the recorded categories or beyond the reported follow-up information.

The record lists no non-serious adverse events for either group: 0 among 19 at risk in the alendronate group and 0 among 17 at risk in the control group. Death counts are not reported in the registry record. The record also does not identify any leading serious adverse-event terms.

Event counts should be read alongside the population at risk, the period of observation, and how events were collected. The registry does not report enough detail about adverse-event definitions, surveillance, follow-up completeness, or attribution to assess those aspects. A zero recorded count is not proof that an event is impossible or that a treatment is free of risk; it means only that no events were recorded in the listed category for the reported group information.

13. What the Result Does, and Does Not, Mean

This registry record describes a study of alendronate added to calcium citrate and vitamin D, compared with calcium citrate and vitamin D alone, in people with postmenopausal osteoporosis or osteopenia. The primary endpoint is percent change in lumbar spine bone mineral density from baseline to month 24. The study's stated purpose also focuses on an experimental computerized analysis of heel radiographic images as a possible way to monitor treatment.

The record does not report numerical results for the primary endpoint or the secondary outcomes. It therefore does not show the direction or magnitude of observed changes, the uncertainty around a group comparison, or whether the image-derived measures were related to changes in bone mineral density. The absence of reported results should not be interpreted as showing no difference or no change.

The study was non-randomized and unmasked. Accordingly, any group comparison would require caution, because the record does not establish that the groups were comparable at baseline or explain how allocation was determined. The noted missing primary endpoint values, which were not imputed, add another source of uncertainty whose impact cannot be assessed from the registry entry.

The registry can establish the broad design, treatment descriptions, endpoint definitions, and recorded adverse-event counts. It cannot, on its own, establish whether alendronate caused a change in bone density or whether heel texture analysis is useful for monitoring treatment. Those conclusions would require outcome results and fuller information about allocation, analysis methods, follow-up, and missing data.

14. Why This Trial Matters Statistically

The study is statistically notable for bringing together a treatment comparison and an assessment of a proposed measurement method. Its primary endpoint is lumbar spine bone mineral density change, while its stated purpose focuses on computerized analysis of heel x-ray images. The listed secondary outcomes include several bone density sites and a range of radiographic texture features. These are distinct kinds of measures, so a monitoring claim would require evidence about how the image-derived measures behave in relation to clinically relevant change, not merely a list of measured features.

The non-randomized allocation is central to interpreting the comparison. In general, treatment groups formed without random assignment can differ for reasons that also affect outcomes. This makes it difficult to separate a treatment-associated difference from baseline group differences, especially when the record does not report allocation procedures or baseline comparability.

The endpoint missingness also matters: the record says some primary endpoint values were not imputed. Missingness can affect precision and, depending on its causes, the validity of a comparison. Because reasons for missingness and sensitivity analyses are not reported, its influence cannot be judged.

Finally, the registry does not report the endpoint estimates, uncertainty intervals, or analysis plan. The record thus makes clear what the study set out to measure, but it does not permit an assessment of the observed effect or of the performance of the proposed image method. This distinction between a clearly specified research question and an available result is important when interpreting registry information.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main limitation for interpretation is that the registry record does not report numerical results for the primary endpoint or the listed secondary outcomes. Without estimates and measures of uncertainty, the magnitude, direction, and precision of changes cannot be assessed.

Allocation was non-randomized, and masking was absent. The record does not explain how participants entered each group or provide baseline characteristics that could help assess group comparability. Confounding therefore remains an important concern for any treatment comparison, and the record does not say whether statistical adjustment was attempted.

The primary endpoint had missing values in both groups, and the record says they were not imputed. Reasons for missingness, details of follow-up, the analysis population, and sensitivity analyses are not reported. The possible effect of missing measurements on the comparison is therefore unknown.

The registry lists multiple secondary outcomes but does not report a multiplicity strategy or a prespecified hierarchy among them. It also does not report the statistical analysis plan, model specification, interim analyses, or trial-specific sample-size rationale. The planning illustration is not a substitute for those details and is not observed power.

Finally, the record does not report detailed adverse-event collection methods or death counts. These limitations restrict the safety interpretation to the event counts recorded in the registry and do not support broader conclusions about event monitoring or follow-up.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Texture Analysis for Postmenopausal Osteoporosis — NCT00145977.
• Registry record last updated 2018-09-11; status last verified 2018-08.
• Results first posted 2018-09-11.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry describes a completed, parallel, non-randomized, unmasked study sponsored by the University of Chicago, with an enrollment of 36. The recorded study start is 2001-07 and primary completion is 2009-12. The population is described in relation to osteoporosis and osteopenia.

The experimental group received alendronate 70 mg once weekly with calcium citrate and vitamin D; the control group received calcium citrate and vitamin D. The primary endpoint was percent change in lumbar spine bone mineral density from baseline to month 24. Secondary outcomes included bone density at other sites and heel radiographic texture features.

The registry record does not report numerical results for the primary or secondary outcomes. It notes missing primary endpoint values that were not imputed, and records 0 serious adverse events in each group. The non-randomized design and absent outcome results limit assessment of treatment effects and the proposed monitoring method.